When the user was writing into an unitialized extent,
ext4_ext_convert_to_initialize() was not requesting journal write access
before it started to modify the extent tree.   Fix this oversight.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
        newblock = iblock - ee_block + ext_pblock(ex);
        ex2 = ex;
 
+       err = ext4_ext_get_access(handle, inode, path + depth);
+       if (err)
+               goto out;
+
        /* ex1: ee_block to iblock - 1 : uninitialized */
        if (iblock > ee_block) {
                ex1 = ex;
                        ex = path[depth].p_ext;
                        if (ex2 != &newex)
                                ex2 = ex;
+
+                       err = ext4_ext_get_access(handle, inode, path + depth);
+                       if (err)
+                               goto out;
                }
                allocated = max_blocks;
        }
        ex2->ee_len = cpu_to_le16(allocated);
        if (ex2 != ex)
                goto insert;
-       err = ext4_ext_get_access(handle, inode, path + depth);
-       if (err)
-               goto out;
        /*
         * New (initialized) extent starts from the first block
         * in the current extent. i.e., ex2 == ex